I've written about the Federal Reserve a good deal, and I've been critical of its willingness to take private sector risk onto its balance sheet by swapping Treasuries for the increasingly dodgy securities produced during the credit bubble.

The (hopefully defunct) Paulson Plan has provided a come to Jesus moment. Having the Fed lend money against overpriced collateral valuations is infinitely superior having the Treasury purchase those securities outright at inflated prices. While the taxpayer is on the hook in either case, at least when the Fed lends, the taxpayer loses only if the borrower first fails. If the Treasury Secretary overpays for an asset, the original owner books the profit and the taxpayer eats the loss directly. A "quiet bailout" from Dr. Bernanke strikes me as much more equitable than a sellers' market fashioned by Secretary Paulson.

Lending from the Fed even against worthless collateral cannot address undercapitalization of the banking sector. Neither could the Paulson Plan, unless the Secretary were to overpay for bank assets. Removing the ability and incentive to overpay on a permanent sale strikes me as a good thing. Recapitalizing the banking system will require more than easy money from the Fed. But easy money, combined with a bit of temporary forbearance on capital requirements, would help blunt the current panic, while leaving bank stockholders and creditors on the hook before any losses would be taken by taxpayers.

The big hazard of this approach is as it has always been, how to wean the banking system from the mother's milk of Fed largesse. Many banks are insolvent, and those banks need to be identified, reorganized and recapitalized (or else liquidated). That should happen quickly — not zombie style over the course of a decade — but it needn't happen instantaneously and simultaneously. In order to even get started, we need to have a fair endgame, a good approach to reorganizing and recapitalizing banks. I'm already on record as supporting either temporary nationalization (a la the Scandinavian model or AIG), or internal recapitalizations via debt to equity conversions. It's possible that a combination of these two approaches, one that doesn't give government direct control over reorganized banks and includes both internal and public sector capital injections, might be more palatable than either choice alone. That will be the subject of my next post.

In the meantime, I would support a standalone act authorizing the Fed to pay interest on deposits immediately. I would prefer that Congress impose limits on the quantity of deposits on which interest can be paid, to limit the risk and interests cost to taxpayers, but that limit could be quite loose for the moment. This approach has the advantage of getting liquidity into the banking system far more quickly than the Paulson Plan ever could have, and drawing a clear line between the liquidity and capitalization aspects of the plan. It could be implemented immediately by passing the one sentence Section 128 of the Paulson Plan in isolation (although again, I'd prefer to muck it up with a limit on the quantity of paid deposits).

Freed of its balance sheet constraint, the Fed might consider injecting funds into the banking system by purchasing a diversified portfolio of holdings in money market funds that trade in commercial rather than government paper. This would help relieve the stresses in the commercial paper market very directly, and reduce the likelihood of a disorderly adjustment in nonfinancial commercial credit markets.